The Path To 4G Will Take Many Turns

A race is going on right now and it's a heated one. It's a race to the next generation of untethered communications—4G mobile broadband wireless. The prize for reaching the finish line first or even finishing at all could be huge. Yet that finish line keeps moving, as the fourth generation (4G) isn't yet fully defined. In fact, spectrum won't be allocated until 2007. Standards probably won't be complete until the year after that. Right now, though, many teams across the globe are investing a lot of time, effort, money, and talent to make sure that they are in the race.

Any road race has three basic fundamentals: a track, a car, and a crew. The worldwide effort to develop 4G systems and devices could be considered a race. Just as the track itself sets the boundary conditions for the race, standards set the boundaries for telecommunications.

The first topic of this article is standards development for 4G, which ranges from the International Telecom-munications Union (ITU) to national and regional standards development organizations (SDOs). The people who are toiling away day in and day out to make sure their team wins the race—the "pit crew" of 4G research—are the engineers working in industry partnerships, high-tech corporations, and university research labs. The second section of this article will focus on their efforts.

Lastly, every team needs a vehicle to propel them to the final goal—the finish line. In the race to 4G, the race cars are the underlying technologies that enable communications to be more seamless, useful, and powerful. In the last section of this article, we will discuss some of the major enabling technologies that fuel progress in this race. We also will examine the types of engineering and test challenges that face designers of 4G systems and devices.

THE TRACK: SETTING STANDARDS
Ask 10 people what 4G means and you'll get 10 different answers. The more important question is: How do we get there? When looking for guidance or direction, it helps to start at the top. In this case, the top is the ITU. This advisory organization was chartered by the United Nations to provide health and human-safety requirements to national authorities. Those authorities, in turn, are charged with regulating the wireless industry in their respective countries.

The ITU is organized into three main sectors. Each sector is broken up into study groups that carry out the majority of the technical work. All ITU guidelines are developed according to a formal process. The study groups address particular technical "questions," which are technology areas that warrant further research. Once a topic has been sufficiently researched and a decision has been made about how to proceed, the group submits a formal "recommendation." This recommendation is then shared with all of the external ITU partners, such as SDOs and national governments.

Two groups within the ITU specifically engage in helping to define the next generation of mobile wireless (FIG. 1). These two groups include:

Working Party 8F (WP8F) in section ITU-R
Special Study Group (SSG) "IMT 2000 and Beyond" in section ITU-T

WP8F is focused on the overall radio-system aspects of 4G, such as radio interfaces, radio-access networks (RANs), spectrum issues, service and traffic characteristics, and market estimations. The SSG "IMT-2000 and Beyond" is primarily responsible for the network or wireline aspects of future wireless systems including wireless Internet, convergence of mobile and fixed networks, mobility management, internetworking, and interoperability.

The main deliverable of WP8F is Recommendation ITU-R M.1645. This recommendation contains the overall goals for the future development of wireless communications. To illustrate the type of work that WP8F does, here is a brief, paraphrased list of the suggestions that are contained in the recommendation (from www.itu.int/ITU-R):

The framework for 4G systems should fuse elements of current cellular systems with nomadic wireless-access systems and personal-area networks in a seamless layered architecture that is transparent to the user.
Data rates of 100 Mbps for mobile applications and 1 Gbps for nomadic applications should be achievable by the year 2010.
Worldwide common spectrum and open, global standardization should be pursued.

THE CREW: CURRENT R&D
Once a recommendation is made, a lot of work must still be done to figure out how to implement it. The implementation is carried out by several groups, which include SDOs, industry forums, and individual corporations designing next-generation networks and devices. Just as a strong pit crew helps keep a race car performing at its peak, the engineers working in these organizations facilitate technical progress toward the final goal of wireless broadband.

Some of the major SDOs are nonprofit regional or governmental bodies, such as ETSI in Europe, CCSA in China, and the TTA in Korea. 3GPP and 3GPP2 are examples of industry SDOs that develop and maintain standards for current 2G and 3G technologies. To achieve maximum efficiency and interoperability, SDOs work together within the framework of the ITU. The ITU actively solicits input from external organizations through structured communication channels. Within the ITU-T, for example, Recommendations A4 and A6 govern cooperation between the ITU, industry consortia, and regional SDOs.

Alcatel, Ericsson, Motorola, Nokia, and Siemens founded the Wireless World Research Forum (WWRF) in early 2001 (www.wireless-world-research.org). That forum's objective is to formulate visions on strategic future research directions for the wireless field. The timeframe for these reflections is in the range of 7 to 12 years from now. The main deliverables of the WWRF are white papers on emerging-technology topics and its seminal Book of Visions. The white papers focus on topics like ad-hoc networking, Ultra Wideband, smart antennas, and reconfigurable architectures. The Book of Visions is a document that collects ideas on the opportunities and challenges of the future wireless world. It includes operational issues like spectrum policy and security.

In 2003, the WWRF announced an effort to establish linkages and discuss common goals with another major industry forum that is contributing to the development of fourth-generation technology: Japan's mobile IT forum (www.mitf.org). In the mITF's 4G Mobile Communication Committee, work is carried out in two groups. The system subcommittee works with external organizations to clarify 4G standards development at a technology level. In contrast, the application subcommittee studies visions and usage opportunities to help design a roadmap toward new lucrative business markets. Goals and recommendations, such as the features and services expected in 4G mobiles, are collected into the Flying Carpet document. (The updated second version of this document can be found on the mITF web site that was listed previously.)

Many special projects pair high-tech corporations with academia to solve some of the challenges that face designers of 4G networks, devices, and services. For instance, the 4G Radio project is a collaboration between Agilent Technologies, STMicroelectronics, Infineon Technologies, and several other industry and academia partners (www.4G-radio.de). Supported by Medea+ and IWT, this project aims to shorten the development time of advanced chip sets and/or systems on a chip (SoCs). To do so, it will create standardized circuit libraries and reconfigurable digital building blocks. These design tools can be used along with proprietary modules to implement the physical layer of new technologies with innovative modulation schemes.

On the services side, the IST-MoDiS (www.ist-modis.org) and IST-MAESTRO (www.ist-maestro.dyndns.org) projects investigate the feasibility of combining broadcast satellite multimedia content (S-DMB) with terrestrial 3G cellular architectures (FIG. 2). This combination could help spur the convergence of different modes of wireless access and content delivery.

THE CAR: ENABLING TECHNOLOGIES
As the racetrack shows us where we need to go, the driver and crew make sure that we use the tools at our disposal to get there. So what are the tools that we use to get to 4G? Which technologies enable mobile broadband wireless? Because the electromagnetic spectrum is a limited resource, much of today's research is aimed at trying to find methods of using this spectrum in more efficient ways. Three main drivers will enable the migration to next-generation broadband wireless: increased capacity (the number of wireless subscribers is still growing); increased data throughput (to enable services that generate higher revenues per user); and seamless interoperability (to increase user acceptance and ease of use.)

One way to increase system capacity is to implement a multiple-input multiple-output (MIMO) antenna scheme. A wireless system with single antennas obeys Shannon's classical limit for capacity, which can be expressed as C = log2(1+SNR). Ideal capacity therefore increases as the log of the signal-to-noise ratio. MIMO systems, on the other hand, increase capacity linearly with respect to the number of transmit and receive pairs that is used.

A MIMO system can be implemented in a couple of ways. One approach is to utilize diversity on the transmit and receive sides to improve robustness and extend range. Alternatively, a number of unique data streams can be transmitted through parallel spatial channels. Most systems will likely use a combination of these two methods (FIG. 3).

MIMO is just one implementation of smart-antenna technology. To optimize the radiation and reception patterns, smart-antenna systems usually combine multiple antenna elements with intelligent and powerful signal processing. Some smart-antenna areas of research include: spatial processing, phased arrays, adaptive arrays, and digital beamforming. In any smart-antenna system, one must consider the following tradeoff: the increase in quality of service (QoS) and coverage gained versus the increase in infrastructure and terminal cost. The area of adaptive antennas is addressed by the ITU's WP8F in Question ITU-R 224-1/8.

With its natural resistance to multipath fading and its ability to support extremely high data rates, orthogonal frequency division multiplexing (OFDM) is a major candidate for fourth-generation air interfaces. Here's the basic idea behind OFDM: If you take a signal and send it over multiple low-rate carriers instead of a single high-rate carrier, the longer symbol periods lessen or eliminate inter-symbol-interference (ISI) problems. The added benefit of the high data rate is then made possible when the multiple carriers (numbering in the hundreds or even thousands) are modulated with higher-order symbol mapping, such as 8PSK or 16QAM. More bits are then forced through the air in each symbol period.

Several major wireless-equipment manufacturers are investing in advanced OFDM research. According to a recent press release, Motorola Labs (www.motorola.com) announced successful field trials of a combined MIMO-OFDM handset. This handset was able to sustain 20-Mbps data rates with low latency while moving at vehicular speeds. Laboratory simulations suggest a maximum uncoded data throughput of 300 Mbps.

OFDM also is the format of choice for emerging technologies like Ultra Wideband (UWB) and WiMAX from the IEEE. There are many variants of OFDM that are optimized for particular applications and channel conditions. They include coded OFDM, wideband OFDM, flash OFDM, and vector OFDM. The OFDM Forum (www.ofdm-forum.com), which is a consortium of wireless companies and university researchers, has been established to lobby national SDOs for a single, universal OFDM standard.

The vision of 4G is a seamless communications environment. Here, interoperability between different systems and different agencies is critical. Older wireless networks are less capable when voice, data, and video information need to move through the same system. This lack of flexibility drives the need for software-defined radios (SDRs).

With SDRs, many communications functions that were formerly carried out solely in hardware can now be performed in software. By downloading specifications over the air, the software-defined radios can be reprogrammed to transmit and receive signals over a wide range of frequencies. They also can emulate virtually any desired transmission or modulation format. This capability will enable a device to perform well on different types of networks and access systems. At the same time, the device will be able to perform multiple functions simultaneously and upgrade new features as they are developed.

The area of software-defined radios is being actively researched by the ITU WP8F in Question ITU-R 230-1/8. An international, nonprofit forum also has been created to help accelerate the development and deployment of SDR technology. It is appropriately called the SDR Forum (www.sdrforum.org).

What about migration strategies? For entrenched cellular-service providers and major wireless-device manufacturers, the most likely path to 4G is to enhance current 2G and 3G systems with increasingly advanced delivery methods. 3.5G technologies, like 1xEV-DO for cdma2000 and HSDPA for W-CDMA, allow for a 3X to 5X jump in data rate on a bit-per-second-per-MHz basis. NTT DoCoMo of Japan has already announced its plans to develop "3.9G" technology.

At Wireless Japan in July of last year, a senior R&D manager explained how 3.9G will bridge the gap between HSDPA at 10 Mbps and 4G at 100 Mbps (www.itmedia.co.jp). This "super 3G" technology, which is envisioned for deployment some time before 2010, will utilize the same spectrum as W-CDMA. It will employ an all-IP network architecture and orthogonal-frequency code division multiplexing (OFCDM) to reach data rates of 30 Mbps. The transition to 4G would then occur by both increasing the bandwidth to 100 MHz and using variable-spreading-factor OFCDM.

A different migration strategy is in store for the current providers of fixed or nomadic wireless access. The IEEE, an industry SDO based in the U.S., has led the way in developing wireless standards that are optimized for a particular domain: 802.15 for the wireless personal-area network (WPAN), 802.11 for the wireless local-area network (WLAN), and 802.16 for the wireless metropolitan-area network (WMAN). Some implementations of these technologies address the 4G goal of very high data rates:

802.15.3a (Ultra Wideband) with a short-range throughput of 480 Mbps
802.11n (MIMO WLAN) with a medium-range throughput of 100 Mbps
802.16-2004 (WiMAX) with a long-range throughput of 75 Mbps (FIG. 4)

To achieve the true vision of 4G, however, one significant hurdle remains: to add mobility to these systems. There are two promising technologies on the horizon that may address the mobility problem—if not the high data rates—head on. The one that the world is likely to see first is 802.16e, an enhancement to the WiMAX standard. Samsung Electronics is working on a proprietary technology that's very similar to 802.16e. Called WiBro (wireless broadband), it is being developed in conjunction with service provider SK Telecom and the government-funded R&D organization, ETRI.

Further out on the horizon is 802.20, which will support digital-subscriber-line (DSL) data rates and mobility up to train speeds. So far, this standard isn't as well defined as 802.16e. But it does have some industry backing from Flarion Technologies and T-Mobile. The system architecture will most likely follow a layered approach with logical link control (LLC). The physical and medium-access control (MAC) layers will therefore deliver services to an IP-based Layer 3 or switching layer, such as point-to-point protocol (PPP) or multi-protocol label switching (MPLS).

DESIGN AND TEST CHALLENGES
Some interim technologies do enable broadband networks to go mobile or cellular networks to deliver broadband. These technologies don't yet meet the data-rate or mobility goals of a true 4G system as defined by the ITU. Continuing research in these areas has many tangible benefits that pave the way to 4G, however. For example, such research ensures a smooth migration from 3G while helping to define the requirements for next-generation data networks and devices. These requirements—some of which were hinted at already within this article—include the capability to support multiple air interfaces, multiple user interfaces, and reconfigurability.

What does this mean for the device designers who are now performing early research and development? One conclusion is that the trend toward integrating the analog front end won't be slowing down anytime soon. Much of the next-generation software-defined terminal will be packaged onto a single, complex system-on-a-chip or system-in-a-package (SiP). Such a system will combine power amplifiers with filters, antennas, and other components, which behave differently in mixed-mode environments. Digital signal processing (DSP) will be used more and more to compensate for analog impairments in front-end blocks.

In order to verify a system of this complexity, test tools will be needed that link the worlds of RF emulation and simulation, digital system modeling, and hardware verification. Because 4G standards aren't clearly defined, finding the right test tools can be difficult. It's impossible to know which of the plethora of emerging technologies will actually grow into market viability. The major key is to invest in a flexible and scalable test platform that covers a broad range of today's formats, meets your requirements for frequency range and bandwidth, and can address emerging needs quickly with new capabilities.

For systems design, look for an electronic-design-automation (EDA) solution that emulates the entire transceiver chain from DSP to RF. This solution should be flexible enough to provide both transistor-level precision and behavioral model efficiency in simulation. It's also helpful if standard design libraries are available for current technologies like UWB and W-CDMA. Those libraries will ease convergence and integration verification. To stress the performance of hardware-receiver designs, procure a high-frequency signal generator that combines large bandwidth to handle complex waveforms with high resolution for dynamic range. The signal generator also should have the ability to download proprietary signals from software packages like Agilent ADS or MATLAB.

For advanced transmitter design, find a signal analyzer with multiple coherent measurement channels that are calibrated in both time and frequency domains. This feature is critical for designing a smart-antenna system. Analyzers that run on a software engine, such as Agilent's Vector Signal Analyzer, offer great flexibility because the data acquisition is independent of the downconverting hardware front end. Engineers can use a high-performance spectrum analyzer if they need extra dynamic range. For extra bandwidth, use an oscilloscope. See Figure 4 for an example of a WiMAX measurement made using VSA software.

The engineers and researchers who conceive and develop next-generation mobile-broadband wireless devices require innovative test tools. These tools must enable proprietary models to be verified both virtually and on a testbench of real-world hardware. In addition, repeatable correlation must exist between the two environments. Consider your favorite test-equipment provider as part of your pit crew. The provider is there to help you accelerate your progress toward the finish line. Equipment providers follow standards development closely. As a result, they can contribute to new product success by offering accurate tools and useful measurement expertise. The race to 4G is far from over. By being more familiar with how it is run, however, you vastly increase your chances of winning.